PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007

Unraveling DONALD R. HODEL University of California 4800 E. Cesar Chavez Ave. Clinostigma Los Angeles, CA 90022 USA in Samoa [email protected]

Field work has revealed that Samoa is home to three of Clinostigma, the handsome, highly ornamental, quintessential palms of the South and West Pacific. Two species, C. samoense and C. warburgii, occur on ‘Upolu while the third one, C. savaiiense, is found on Savai’i.

Samoa includes nine inhabited islands and two great, sweeping arc from the Ogasawara and political regions in the Pacific Ocean between Caroline Islands in the west Pacific Ocean to 13 to 15° south and 168 to 173° west, about New Ireland, , Vanuatu, Fiji, 4200 kms south of Hawaii, 2200 kms west of and Samoa in the south Pacific. While unusual, Tahiti, and 2900 kms northeast of New this vast, insular distribution is not unique Zealand. The two largest islands, Savai’i and among the palms. The coconut, Heterospathe, ‘Upolu, comprise the independent state of Metroxylon, and Pritchardia, among others, have Samoa (formerly Western Samoa) while Tutuila a similar distribution. Clinostigma are highly and several smaller islands in the eastern part ornamental and typically have a handsome, of the archipelago comprise the U. S. territory smooth, waxy-white, ringed trunk topped with of American Samoa. Polynesians settled and a conspicuous, lime-green crownshaft and inhabit the islands and enjoy a year-round, splendid crown of spreading with wet, tropical climate. Centuries of human elegantly pendulous pinnae. activity and a series of strong tropical cyclones Samoan Clinostigma differ from other members in the last 20 years devastated much of Samoa’s of the on nearby island groups, such as fragile forests. Undisturbed, closed-canopy C. exorrhizum (Fiji) and C. harlandii (Vanuatu), primary forest is uncommon while highly in the lack of prominent stilt roots at the base disturbed and altered primary and secondary of the trunk and black rather than red . forests, typically with many exotic, invasive weed species, predominate. In Samoa Clinostigma has had a long and somewhat tortured and complex taxonomic Six genera of palms are recorded for Samoa: history comprising up to six taxa and eight Balaka (2–4 species); Clinostigma (3 species); names. In December 1978 I visited Samoa and Cocos nucifera (coconut); the mysterious and made four collections of of Clinostigma enigmatic Drymophloeus whitmeeanus; from ‘Upolu, two each of what I called C. Metroxylon (1 or 2 species); and the rather shy samoense and C. onchorhynchum, and Solfia samoensis. All genera and species occur distributed them to several botanical gardens in Samoa while only the coconut occurs in in Hawaii and palm growers in Hawaii, American Samoa (Clinostigma and Metroxylon California and Florida. At the time, I based have been recorded from Tutuila although it these species determinations entirely on is thought that they are cultivated and were shape and size and position of the stigmatic brought from Samoa). remains because the most current information The focus of this paper is on Clinostigma, which in the literature at that time (Moore and includes about 11 or so species native in a Fosberg 1956, Moore 1969) and personal

PALMS 51(1): 11–29 11 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007 communication with the late Dr. Harold E. Botanical Garden, and a subsequent visit in Moore, Jr. suggested that these were the most 2005, though, prompted me to investigate important and reliable characters for further Clinostigma on ‘Upolu. Based on distinguishing species of Clinostigma. architecture alone, there are clearly at least two taxa, but the appropriate Near sea level along the northern and southern application of names to these taxa remained coasts of ‘Upolu I made collections in two elusive. localities where the fruits were rounded, 7–10 mm diam., and with the stigmatic remains In October 2005 I again visited Samoa and about midway between the top and bottom. I other islands in the southwestern Pacific to referred to these two collections as Clinostigma gather information and take photographs as samoense. In the central highlands of ‘Upolu I part of a project I am leading that will result made collections in two localities where the in a publication on the palms of Pacific Islands. fruits were markedly longer than broad, 15–18 This paper summarizes the findings about mm long, and with the stigmatic remains near Clinostigma in Samoa from that trip as well as the top. I referred to these two collections as information gleaned from the cultivated C. onchorhynchum. in Hawaii that were grown from seeds I At least one each of my Clinostigma collected in Samoa in 1978, examination of onchorhynchum and C. samoense collections specimens in several herbaria, and a review of from ‘Upolu have been flowering and fruiting the literature. for several years at Ho’omaluhia Botanical History of Clinostigma in Samoa Garden of the Honolulu Botanical Gardens system in Hawaii. While visiting Ho’omaluhia The tortured and complex taxonomic and in 2004, I examined these collections carefully nomenclatural history of Clinostigma in Samoa and was able to distinguish the two taxa based begins and is centered on the island of ‘Upolu solely on fruit size and shape, as I had done (Table 1). Hermann Wendland (1862) more than 25 years earlier on ‘Upolu in Samoa. established Clinostigma when he named and However, I was surprised to notice that the described C. samoense from material that branching pattern or architecture of the Pickering (Pickering s. n.) had collected on of the two taxa was markedly ‘Upolu during the United States South Pacific different as well, a character to which I and Exploring Expedition of 1833–1842. Asa Gray others had given little, if any, attention. The of Harvard University, who was preparing the C. onchorhynchum, with larger fruits longer botanical account of the Expedition, had than broad, had broom-like inflorescences forwarded Pickering’s material to Wendland, branched to two orders and thick, coarse, one of the leading palm botanists of the time. narrowly diverging rachillae. The C. samoense, According to Christophersen (1935) and with smaller, rounded fruits, had diffuse Moore and Fosberg (1956), Wendland had inflorescences branched to three orders and commented that the material upon which he slender, widely spreading rachillae. based the new genus and species was My surprise at the differences in inflorescence incomplete and contained fragments from architecture between what I and others had Savai’i. Wendland’s specimen consisted of part referred to as and C. of a , one of the main branches of an onchorhynchum was due in part to the fact that inflorescence with fertilized pistillate , Cox and Moore (1986) had thoroughly and the tips of some rachillae with immature discounted fruit shape and size as reliable fruits, the latter labeled “Savai’i” and with the characters for distinguishing these two species. stigmatic remains midway between the base Indeed, in their paper, they relegated C. and apex (Christophersen 1935). Although onchorhynchum to a synonym of C. samoense, Wendland’s material was incomplete and a fact to which I was aware during visits to perhaps mixed, Moore and Fosberg (1956) felt Hawaii in 1998 and 2000. At those times I that “the major part of the description informed the staff at Ho’omaluhia Botanic corresponds with what appears to be a Garden about this nomenclatural change and, duplicate specimen at the Gray Herbarium” based on my recommendation, they adjusted (GH). the labels accordingly. The specimen at GH consists of part of a leaf, The rather dramatic differences in a branch of the inflorescence, and loose inflorescence architecture I had noticed immature and rounded, nearly mature fruits, between the two taxa in 2004 at Ho’omaluhia all labeled as being from ‘Upolu and annotated

12 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007

i ) i ’ ’ i d i

a e a a , , t v i v o o o a ’ ’ c a

S o o S m

y t t , a , t o o i S u a l

n n o n n a n l a a a o c r a u i L L v u u l e o t

g l l l t a o a e e

s t u o o c k k p o e a a p p a a o n U U ( L ‘ L L W M M U 7 6 2 d 2

e t . n . i e n n c 2 s

. r . 2 s 6 e s n ? 3

? 4 h

o e g i ? e p 2 e

t n

k l o i l c c t m e l r e t s e e e p i l i l n k r u h w i c o h a e i o C P R W P C V

. ) ) 6 6 9 0 0 9 0 0 9 2

2 1 ( (

o

; ) t l l ) )

0 e e 6 5 2 1 d d 8 3 6 9 o o ) 9 9 8 1 6 1 H H

1 ) 1 ( ,

0

(

; ; ;

; ; 2

, 7 ; ) ) ) ) ) ) 0 ) ) e s 6 ) n 0 r e 8 2 2 2 2 2 2 6 6 ) ) ) ) 8

5 t u e 9 o ) ) ) 9 9 9 9 9 9 , ) ) 7 7 0 4 3 3 t s a 1 3 1 o 8 5 5 8 9 9 9 9 9 r 9 6 6 9 9 ( 1 3 1 1 a ( 9

D e t 2 3 3 1 1 1 1 1 1

9 0 0 1 1 9 9 9 9

1 ( M ( ( ( ( ( r s d ( ( 9 9 9 h 9 0 0

( 1 1 1 1

e

& r r r r r ( ( ( ( n 1 1 1 r s s 1 2 2 p g d i

g i i ( ( ( i ( ( ( e e e e e r a l i i i i r o

l l l l l

l n l e r r r r o o n t t t t l l l u t t t t t o l l i e a d s a a a a s s s s s e e e e e e h

b t i g g i i i i i h r r r h c c c c t r n r r d d d x t

r r r c c c c c n n h h h h h e a a o o o o h u e e u e e e u u d a a H A W R L B B L W C M W H W W B H W B B C B W B n a

a ” o e e u m s l s m u n o n a e h e i e p S c i s o

s n a i U m n e e e ‘ n y s v m e s s s u

r i h n a a

o i n n n h n e r ) ) e s s l e e e c r t

o e o m n n o o o s e n a a h p e a t e o o y m c s s s m m m i i u m m

h a t t n n a a a a a r g g W s a a a e o s s s i i

v E

o

t t

o n n “ a m s s n h a a a a i i a

i i g c . o o t m z i b b i m m m m n t n n p a n a i i g g g g s o s s h e l l m m i i i i

r o k t t t t m r o o a a a C C s s s s n o

g o c c i

o o o o f f i l ( ( h m m m s s s s s s x

t n n n n o o p g g g e e e e e e C E i i i i s e e

i i i i i i i i i l y l l l

t t t o c c c c c c s s s n n C C C C C m m m m e e e e e e n

o o o a a o o i y y f f f f f p p p p p p

l n n n s s s s s s s n n i i i o o o o o n n d d

l l l

C u . . . o o . .

e e t C C C w w w w w w w w f s s n n n n n n n

a e e e e e e e e a a s s s o t y y y y y y y

a a s S a s n n s n n n n n n s s s b b s s e m m a u n h m

c f u s n e i i e o n s i y s

s i a g n h n i i y n r r e l e l r e i l u e o i o i o e a i b p h l a r m c u w e v m m a a o n p a a a s o s s w p v u

m

a a a a a a a a u i v t

S m m m m m m

n a . g g g g g g e z i i i i i i 1 i k t t t t t t

s s s s s s h o e e r o o o o o o l h r m n n n n n n p b o i i i i i i a l y l l l l l a x T N C C C C C C E C

13 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007

1. Holotype of Clinostigma samoense (GH). with an unpublished name in Wendland’s specimen. The inflorescence is distinctive in its handwriting (Moore and Fosberg 1956). The broom-like appearance with thick, coarse, name “Savai’i” does not appear on the narrowly diverging rachillae (Fig. 1).

14 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007

Over 35 years later, Warburg (1898) examined thus, he was transferring it to Cyphokentia and material that Reinecke (Reinecke 322) had proposing the new combination? collected at Lake Lanoto’o in the central highlands of ‘Upolu in 1894. Warburg One might argue that Warburg was, indeed, compared it to Wendland’s Clinostigma naming a new species because he cited “n. samoense, and, while he stated that Reinecke 322,” obviously referring to Reinecke 322, and 322 was probably referable to C. samoense, he enumerated several differences between the also enumerated several differences between two. If so, then two different types are involved the two. Warburg then apparently proposed and Cyphokentia samoensis must be a new the new name Cyphokentia samoensis although species name, not a combination, and the his intentions are far from clear. names are considered heterotypic (based on different types). It is unfortunate, though, that His actions raise these questions: Had he Warburg chose the epithet “samoensis” for decided that Cyphokentia samoensis, based on Reinecke 322 because it offers only confusion, Reinecke 322, was distinct from Clinostigma not clarity, to the situation. samoense and, thus he was naming a new species; or, had he decided that Clinostigma However, I feel that Cyphokentia samoensis was an incorrect genus for this species and, must be regarded as a new combination, based

2. Holotype of Clinostigma onchorhynchum (K).

15 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007

3. Holotype of Clinostigma powellianum (K).

16 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007 on Clinostigma samoense, rather than a new the type, similar to those of C. samoense, are species because Warburg used the epithet mostly rounded with the stigmatic remains “samoensis” and he did not use the midway between the base and apex or toward abbreviation “n. sp.” (new species) after the the apex. Strangely, Beccari’s illustration in the new name or provide a Latin description, as he literature clearly shows the fruits to be longer did for other, obviously new species he named than broad with the stigmatic remains very in the same article. Because he was not naming near the apex. a new species, he was not designating a type Beccari’s other new species, Clinostigma when he referred to Reinecke 322; thus, the powellianum (C. “powelliana”), was based on names Cyphokentia samoensis and Clinostigma material without a specific locality in Samoa samoense must be considered homotypic that Powell (Powell 246) had collected. Beccari (based on the same type). noted it had more spreading inflorescences (he Regardless, Warburg’s illustration of used the term “diffuses”) branched to two inflorescences and a branch of the orders and nearly rounded fruits. Examination inflorescence with narrowly diverging rachillae of Powell’s type material at K, though, shows appear broom-like and somewhat similar to the thick, coarse rachillae to be narrowly that of Clinostigma samoense but the rachillae diverging and scarcely different from those of are much more slender. (We shall see later that C. samoense and C. onchorhynchum (Fig. 3). a duplicate of Reinecke 322 discovered at the Fruits, very similar to those of C. Bishop Museum, Honolulu, Hawaii (BISH) has onchorhynchum and even C. samoense, are slender rachillae like that in the illustration, mostly rounded with the stigmatic remains but they differ dramatically in their spreading near the apex. nature.) Fruits of Reinecke 322, as depicted in Some years later, Burret (1928), after examining Warburg’s description and illustration, are Rechinger’s collections from Lake Lanoto’o, immature with the stigmatic remains near the concluded that Cyphokentia samoensis was best apex. placed with Clinostigma onchorhynchum, not Several years later, Rechinger (1907, 1910) C. samoense, as earlier workers had suggested listed Cyphokentia samoensis from ‘Upolu and or concluded. cited two of his collections from higher Later, Beccari had apparently become skeptical elevations in the central part of the island, about listing Cyphokentia samoensis as a including Lake Lanoto’o, and noted it also synonym of Clinostigma samoense. In a occurred on Savai’i. Rechinger stated that posthumously published paper that Martelli Clinostigma samoense was probably a synonym prepared after Beccari’s death in 1920, Beccari of Cyphokentia samoensis. (Martelli 1934) reversed himself and Shortly thereafter Beccari (1910), after resurrected Cyphokentia samoensis from examining the sparse material composing synonymy with Clinostigma samoense and Reinecke 322 at Breslau, took the opposite renamed it Clinostigma warburgii (the epithet position and listed Cyphokentia samoensis as a samoense already being taken by Wendland), synonym of Clinostigma samoense, indicating stating it differed sufficiently in the nature of that he thought the two were heterotypic and the fibers of the fruit mesocarp. Warburg was naming a news species in 1898. Obviously, Beccari’s interpretation of Warburg’s Three years later Beccari (1913) named and intentions when he named Cyphokentia described two species of Clinostigma from samoensis over 35 years earlier differs from Samoa, both with apparently mature fruits, mine. Although he provided a Latin the first known for the genus. One, C. description, Beccari used the abbreviation “n. onchorhynchum (C. “onchorhyncha”), was based nov.” (new name), indicating he was not on material Whitmee (Whitmee s. n.) had naming a new species but simply proposing a collected at Lake Lanoto’o on ‘Upolu in 1875. new name for Warburg’s existing species. Thus, Beccari noted it had broom-like inflorescences Beccari clearly regarded Cyphokentia samoensis (he used the term “scopaeformis”) branched to and Clinostigma samoense as heterotypic, two orders. While Whitmee’s sparse, believing when Warburg named Cyphokentia incomplete type at Kew (K) does not show the samoensis he did so as a new species rather broom-like architecture, the rachillae are thick, than a new combination. Also, there was no coarse, and narrowly diverging (Fig. 2) and are need for Beccari to designate a type formally rather similar to those of C. samoense. Fruits of because a “n. nov.” is typified by the name it

17 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007

4. Isotype of Clinostigma warburgii (BISH).

replaces, which in this case is Cyphokentia relatively minor point because he had clearly samoensis; thus, under Beccari’s interpretation, come to the correct and appropriate decision the type for Clinostigma warburgii is that Reinecke 322 was distinct from Wendland’s automatically Reinecke 322. Clinostigma samoense. Whether Beccari employed the terms “n. nov.” or “n. sp.” in However, some might contend, as I do, that, formalizing his position is a minor, irrelevant if Warburg (1898) were not naming a new detail, especially because rules governing such species but simply making a new combination protocol were lacking and unclear at that time. when he proposed the new name Cyphokentia What is important is Beccari’s intent, which is samoensis, then Beccari would have been clear, and it should render moot whether obliged to name Clinostigma warburgii as a new Warburg was naming a new species or simply species and state “n. sp.,” rather than “n. nom.” making a new combination when he proposed as he had done, and designate a type, because Cyphokentia samoensis. there would have been no existing species and type upon which Beccari could have based his Martelli (1935), in a listing of palm genera and new “n. nom.” species, followed Beccari and continued to Although I feel that Beccari erred, his recognize Clinostigma warburgii but placed C. misinterpretation of Warburg’s intent is a powellianum as a synonym of C. samoense.

18 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007

5 (left). Clinostigma samoense has broom-like inflorescences branched to two orders with thick, coarse, narrowly diverging rachillae, Ho’omaluhia Botanical Garden (HBG 78.0891) ex Tiavi, ‘Upolu, Samoa (Hodel 469). 6 (right). Clinostigma warburgii has “bushy” inflorescences branched to three orders with slender, widely spreading rachillae, Ho’omaluhia Botanical Garden (HBG 78.0889) (Hodel 2008) ex Salelesi, ‘Upolu, Samoa (Hodel 462).

Christophersen (1935) named and described For example, Christophersen (1935) felt that the first Clinostigma from Savai’i, C. savaiiense, Pickering’s and Reinecke’s material had based on material he had collected differences at least of specific value because (Christophersen & Hume 2267) above Matavanu the fruits, while immature in both cases, were Crater. Nearly simultaneously, Burret (1935a) different, rounded in Pickering’s and named and described another species from somewhat oblong in Reinecke’s. Moore and Savai’i, Exorrhiza vaupelii, from material that Fosberg (1956) and Moore (1969) stated that Vaupel had collected near Maugaloa Clinostigma onchorhynchum and C. savaiiense, (“Mangaloa”), and almost immediately with fruits having apical stigmatic remains, suggested transferring it to Clinostigma (Burret differed at a specific level from C. samoense, 1935b). with fruits having stigmatic remains midway between the base and apex. Although the naming of new species of Clinostigma from Samoa ceased after 1935, In contrast, Cox and Moore (1986) lost favor subsequent workers continued to interpret and with fruit size and shape as reliable, reinterpret these earlier works and meager distinguishing characters and, after examining collections. Despite the fact that many of the collections with immature fruits from several earlier collections had immature fruits, later localities on ‘Upolu, relegated Clinostigma workers placed much emphasis on fruit size onchorhynchum to synonymy with C. samoense. and shape and the position of the stigmatic Unfortunately, they did not address the remains, either in the middle or towards the disposition of any other names of Samoan top of the fruit, in distinguishing species. Little Clinostigma. or no significance was given to inflorescence Whistler (1992), returning to fruit size and architecture. shape as reliable, distinguishing characters,

19 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007 stated that there were two species on ‘Upolu, While fruit shape and size and, to some extent, C. onchorhynchum (with large oblong fruits) rachilla size, by themselves are helpful in and C. samoense (with small rounded fruits), distinguishing the three species, they can vary and one species on Savai’i, C. savaiiense (with a little. Thus, it is better to rely on inflorescence small oblong fruits). He listed C. warburgii and architecture and the presence or absence of C. powellianum as synonyms of C. samoense lepidia on the abaxial surface of the pinnae and C. vaupelii as a synonym of C. savaiiense. for identification. Unfortunately, this information, especially about inflorescence A few years later in a very short, popular architecture, is lacking in nearly all collections summary of the genus (Hodel 1999), I of Samoan Clinostigma. As was typical of many recognized only one species in Samoa, palm collections of the era, the early Clinostigma samoense, and listed all the other collections of Samoan Clinostigma, including names from Samoa as synonyms. types, are rather poor. Many recent collections Recently, after examining Clinostigma in are of similar quality. They are meager, sparse, Samoa, my living collections in Hawaii, types and incomplete, often lacked mature fruits, of C. samoense, C. onchorhynchum, and C. and label data and other information, powellianum, and the literature, I recognized especially about diagnostic characters of three species in Samoa based entirely on taxonomic value that are difficult or inflorescence architecture, two on ‘Upolu and impossible to encompass on the herbarium one on Savai’i (Hodel 2006). On ‘Upolu I sheet, are scant or non-existent. recognized C. samoense (inflorescence branched to two orders with thick, coarse, Key to Species of Samoan Clinostigma narrowly diverging rachillae) and C. sp. 1. Inflorescences broom-like, typically “Eastern ‘Upolu” (inflorescence branched to branched to two orders, rarely branched to three orders with slender, widely spreading three orders; rachillae thick, coarse, narrowly rachillae). I considered C. onchorhynchum, C. diverging ...... 1. C. samoense powellianum, and C. warburgii as synonyms of C. samoense. I recognized C. savaiiense and its 1. Inflorescences not broom-like, typically synonym C. vaupelii as restricted to Savai’i. branched to three orders; rachillae slender, widely spreading ...... 2 However, the recent discovery of authentic material of Reinecke 322 (Fig. 4) at BISH was 2. Abaxial surface of pinnae moderately key to unraveling the Clinostigma puzzle in covered with gray to brownish, circular to Samoa because I could clearly see that, with its nearly linear lepidia; fruits typically oblong . . slender, widely spreading rachillae, it ...... 2. C. savaiiense represented Clinostigma sp. “Eastern ‘Upolu”; thus, the best name for this species is C. 2. Abaxial surface of pinnae lacking lepidia; warburgii. fruits typically +/- rounded . . . . 3. C. warburgii Summary of Clinostigma in Samoa 1. Clinostigma samoense H. Wendl., Bonplandia 10: 196. 1862. Cyphokentia In summary, there are clearly three species of samoensis (H. Wendl.) Warb., Bot. Jahrb. Syst. Clinostigma in Samoa, two on ‘Upolu and one 25: 588. 1898. Type: Samoa, ‘Upolu, Pickering on Savai’i, and we can now apply names to s. n. (holotype GH). them with some degree of confidence. On ‘Upolu, C. samoense has broom-like Clinostigma onchorhynchum Becc., Webbia 4: inflorescences branched to two orders with 284. 1914. Type: Samoa, ‘Upolu, Whitmee s. thick, coarse, narrowly diverging rachillae (Fig. n. (holotype K; isotype FI). 5) and typically larger, oblong fruits with the Clinostigma powellianum Becc., Webbia 4: stigmatic remains toward or near the apex. 286. 1914. Type: Samoa, Powell 246. Also on “Upolu, C. warburgii has “bushy” (holotype K; isotype FI). inflorescences branched to three orders with slender, widely spreading rachillae (Fig. 6) and Solitary, unarmed, pleonanthic, monoecious, typically smaller, rounded fruits with the forest palm 15–20 m tall (Fig. 7). Trunk erect, stigmatic remains near the middle. On Savai’i, green with white waxy glaucous indument, C. savaiiense has open inflorescences sparsely aging to grayish white or brownish, ringed, branched to three orders with slender, 25–30 cm diam., internodes 2.5–7.5 cm. Leaves spreading rachillae, smaller oblong fruits, and 15, ascending to spreading; sheath 1.5 m long, abaxial surface of pinnae moderately covered lime green with slight glaucous bloom, with small scales (lepidia). forming a conspicuous and prominent

20 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007

7. Clinostigma samoense, Tiavi, ‘Upolu, Samoa.

crownshaft; petiole 30 cm long, concave complex distally until most distal represented adaxially, rounded abaxially; rachis 2.6–2.8 m by up to 15 simple rachillae; peduncle 10–15 long, flat adaxially, rounded abaxially; pinnae cm long, 20–25 cm wide at base and ± bulbous up to 75 per side, elegantly and pendulous, and swollen, 7 cm wide and 3–4 cm thick at regularly arranged, lower middle the largest, to prophyll scar, light green with glaucous bloom; 65–85 × 3–4 cm, proximal pinnae to 30 × 1 cm, prophyll 1.5 m long, equaling crownshaft, distal pinnae to 20 × 1.5 cm, prominently 3- attached 4–5 cm above base; peduncular bracts nerved adaxially, tan medifixed ramentae to 1 2, attached 1 and 3 cm respectively distally of cm long on prominent midrib abaxially near prophyll attachment, short, rudimentary, 1–3 rachis, otherwise glabrous. Inflorescences 6–9, cm long, triangular, brown, typically caducous; infrafoliar, to 1.3 × 0.9 m, broom-like, rachis 20–43 cm long with simple rachillae branched to 2 orders (rarely branched to 3 distally and up to 15 branches proximally, orders and then with more moderately most proximal first order branches with diverging branches and rachillae), most unbranched portion 6.5–15 cm long, rachis of proximal branches the largest and most first order branches 5–15 cm long; rachillae complex with up to 7 rachillae each, branches 45–75 cm long, white in , greenish in becoming progressively smaller and less fruit, narrowly diverging, 1.0–2.5 mm diam.

21 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007 distally with staminate flowers only and there staminodes 6, 1 mm long, tooth-like, acute, slightly filiform, attenuate and ± flexuous, 3–5 4–6 toward side of fruit with stigmatic beak. mm diam. proximally with triads and later Specimens Examined. SAMOA. Powell 246 fruits, coarse; proximal branches subtended by (holotype of Clinostigma powellianum, K; brown, long-lanceolate bracts to 60 cm long, isotype FI); ‘Upolu: Pickering s. n. (holotype typically caducous, more distal branches and GH); Malololelei, 550 m elev., 17 Aug. 1929, rachillae subtended by rudimentary bracts 1–3 Christophersen 303 (BISH); Lake Lanoto’o, cm long. Flowers borne in triads in cleft-like Whitmee s. n. (holotype of Clinostigma depressions in proximal one-half to two-thirds onchorhynchum, K; isotype FI); 700 m elev., 22 of rachillae, solitary or paired staminate flowers Aug. 1929, Christophersen 383 (BISH); 24 June only in distal one-third to one-half of rachillae, 2001, Whistler 11532 (HAW); 29 Aug. 2004, each triad subtended proximally by a lip-like Whistler 11791 (HAW); Tiavi, 600 m elev., 13 bracteole, triads 3–6 mm distant in 2 spiraling June 1976, Whistler 3506 (BISH); 740 m elev., rows; floral bracteoles 5–6, unequal in size, 3 16 July 1971, Moore 9978 (BH); Magiagi, 600 inner ones broadly triangular to crescent- m elev., 27 Aug. 1991, Whistler 8117 (HAW); shaped, 0.3 × 1.5 mm, imbricate, 3 outer ones Afiamalu, 510 m elev., 25 Sept. 1991, Whistler often more prominent, the 2 outer lateral ones 8388 (HAW); west of Afulilo, 300 m elev., 30 0.5 × 0.4 mm, tooth-like, outer middle July 1977, Whistler 3873 (HAW); E of Afulilo, (proximal) one especially conspicuous and 275 m elev., 30 July 1977, Whistler 3875 exceeding the triad bracteole, triangular, 1 × (HAW); 5 km E of Afulilo Dam, 450 m elev., 1.3 mm with prominent briefly acute apex, 17 May 1996, Whistler 10036 (HAW). white. Flowers not seen. Fruits 10–20 × 7–13 mm, typically oblong, less frequently Distribution and Ecology. Clinostigma samoense somewhat rounded, black or purplish black primarily occurs in the central highlands of with glaucous bloom, stigmatic remains near ‘Upolu in the districts of East and West Faleata, apex, beak-like; perianth 5 × 10 mm, cupular; West Vaimauga, Safata, and Si’umu, from 300 sepals 5 mm high, imbricate nearly to acute to 800 m elevation in moist to wet, usually apex, broadly rounded, petals 5 mm high, disturbed, often open forest. It has also been imbricate nearly to apex, broadly rounded; collected a few times in eastern ‘Upolu near

8. A possible hybrid of Clinostigma samoense and C. warburgii, which might be C. powellianum, is this at Ho’omaluhia Botanical Garden (HBG 78.0891). It has more spreading inflorescences branched to three orders but with thick, coarse, moderately diverging rachillae.

22 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007

Afulilo Dam in the district of Va’aofonoti. species, Clinostigma samoense and C. warburgii, Typically, C. samoense occurs as scattered, together, making a case for just one highly emergent individuals on steep slopes or in variable species. However, there does not more level areas. appear to be a continuum of variation with a multitude of intermediate forms from one To a great extent altitude and location on species to another, which one would expect if ‘Upolu separate the ranges of Clinostigma there were just one highly variable species. samoense and C. warburgii. They do overlap Rather, there are the two distinct species with somewhat in the central highlands near Lake one additional taxon more or less exactly Lanoto’o and apparently again near Afulilo in intermediate between the two with no other the eastern part of the island. The altitudinal variation present. Perhaps future study distribution of the two species is dramatically employing DNA will be able to sort these taxa demonstrated along the southern half of the out more satisfactorily. cross-island road from Apia to Si’umu. Near Tiavi at the summit at about 800 m elevation Several of Whistler’s Clinostigma collections and down to about 600 m elevation one finds from Upolu housed at HAW are difficult to C. samoense. From about 600 m down to sea assign to species because they are incomplete, level at Si’umu one finds C. warburgii. The two consisting only of pieces of fruiting rachillae, species are easy to distinguish from the auto which, by themselves, are hardly diagnostic. as one passes along the road because of the These include Whistler 3873, 3875, and 10036 differences in inflorescence architecture. The from Afulilo, which is in the middle of the changeover from one species to the other range of C. warburgii, and Whistler 8387 and occurs rather quickly around 600 m elevation. 8388 from Afiamalu, which is in the middle of the range of C. samoense. Based on fruit shape That their ranges briefly overlap raises the alone, I have tentatively included the possibility of hybrids. Indeed, I suspect that collections from Afulilo and Whistler 8388 with Clinostigma powellianum might actually be a C. samoense and Whistler 8387 with C. hybrid, and its placement with C. samoense is warburgii. somewhat problematic. It has the thick, coarse rachillae of C. samoense but Beccari described While Whistler (1992) was correct in the inflorescence as spreading (he used the determining that two species of Clinostigma term “diffuses”) and possibly branched to three occurred on ‘Upolu and could be distinguished orders. Also, he did not use the term by fruit size and shape, what he referred to as “scopaeformis,” meaning broom-like, as he had C. onchorhynchum (with large oblong fruits) is actually C. samoense and what he referred to done for C. onchorhynchum. Unfortunately, the as C. samoense (with small rounded fruits) is meager nature of Powell’s type tells us nothing actually C. warburgii. about the inflorescence architecture. However, one plant out of about 25 cultivated at The palm illustrated in Langlois (1976, fig. 49, Ho’omaluhia Botanical Garden in Honolulu p. 47), captioned as Clinostigma oncho- originating from my collection of C. samoense rhynchum, is actually C. samoense. (broom-like inflorescence branched to two 2. Clinostigma savaiiense Christoph., Bernice orders; thick, coarse, narrowly diverging P. Bishop Mus. Bull. 128: 28. 1935. Type: rachillae; elongated fruits) in December 1978 Samoa, Savai’i, Matavanu Crater, (HBG 78.0891) has more spreading Christophersen 2267 (holotype BISH; isotypes inflorescences branched to three orders with K, US). thick, coarse, moderately diverging rachillae (Fig. 8). Moore 9978 (here referred to C. Exorrhiza vaupelii Burret, Occas. Pap. Bernice samoense) and 9983 (here referred to C. P. Bishop Mus 11(4): 4. 1935. Clinostigma warburgii) also depict this hybrid nature in vaupelii (Burret) Burret, Notizbl. Bot. Gart. their inflorescence architecture. In both the Berlin-Dahlem 12: 593. 1935. Type: Samoa, cultivated plant in Honolulu and Moore’s Savai’i, Maugaloa, Vaupel 605 (holotype B?). collections, the inflorescences are spreading Solitary, unarmed, pleonanthic, monoecious, because of the wider angle of the proximal forest palm 10–20 m tall (Fig. 9). Trunk erect, primary branches. In all case the secondary green with white waxy glaucous indument, branches and rachillae of the primary branches aging to grayish white or brownish, ringed, are only moderately diverging. 15–25 cm diam., expanding to 45–55 cm diam. Some might contend that this hybrid is at base and their supported with prominent actually an intermediate form that ties the two prop roots to 60 cm long and 1.0–1.5 cm

23 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007

9. Clinostigma savaiiense, Matavanu, Savai’i, Samoa.

diam., internodes to 10 cm. Leaves 10–15, arranged, lower middle the largest, to 55–80 × ascending to spreading; sheath 75 cm long, 2.2–4.0 cm, slightly falcate, tips splitting and lime green with slight glaucous bloom, becoming tattered, prominently 3-nerved forming a conspicuous and prominent adaxially, tan medifixed ramentae to 1 cm long crownshaft; petiole 30 cm long, concave on prominent midrib abaxially near rachis, adaxially, rounded abaxially; rachis 1.8–2.1 m pinnae moderately covered abaxially with long, flat adaxially, rounded abaxially; pinnae brownish, circular to nearly linear, raised up to 55 per side, slightly pendulous, regularly lepidia or wart-like structures to 0.4 mm long

24 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007 arranged in raised lines parallel to pinnae rounded-acute apex, finely ribbed; pistil 2.5 × margins. Inflorescences 5–6, infrafoliar, to 1 × 1.5 mm, ovoid. Fruits 13–15 × 7–9 mm, 0.8 m, branched to 3 orders (Fig. 10), most oblong, black, stigmatic remains near apex, proximal branches the largest and most beaklike; perianth 4–5 × 7 mm, cupular; sepals complex, branches becoming progressively 3 mm high, imbricate nearly to apex, broadly smaller and less complex distally until most rounded, margins thin, petals 4–5 mm high, distal represented by simple rachillae; peduncle imbricate nearly to apex, broadly rounded; 5–15 cm long, 15 cm wide at base and ± staminodes 6, 1 mm long, triangular, acute. bulbous and swollen, 4 cm wide at prophyll Specimens Examined. SAMOA. Savai’i: scar, light green with glaucous bloom; prophyll Matavanu Crater, 1300 m elev., 24 Sept. 1929, 1 m long, equaling crownshaft, attached 10 Christophersen 808 (BISH); 900 m elev., 10 July cm above base; peduncular bracts 2, attached 1931, Christophersen & Hume 1946 (BISH); 15 2 and 4 cm respectively distally of prophyll July 1931, Christophersen & Hume 2078 (BISH); attachment, not seen, typically caducous; 5 Aug. 1931, Christophersen & Hume 2266 rachis 75 cm long with simple rachillae distally (BISH); Christophersen & Hume 2267 (holotype and up to 15–21 branches proximally, most BISH; isotypes K, US); Christophersen & Hume proximal first order branches with unbranched 2273 (BISH); 680 m elev., 23 July 1971, Moore portion 10 cm long, rachis of first order 9982 (BH); above Sala’ilua, 1400 m elev., 8 branches 30 cm long; most proximal second Sept. 1931, Christophersen 2565 (BISH); 6 Nov. order branches with unbranched portion 8 cm Sept. 1931, Christophersen 3088 (BISH); W of long, rachis of second order branches 5 cm Mata-ole-Afi, 1500 m elev., 31 May 1975, long; rachillae to 45 cm long, white in flower, Whistler 2564 (BISH, HAW). greenish white in fruit, diffuse, spreading from rachises at angles of 45–90 degrees, 0.8–1.0 Distribution and Ecology. Clinostigma mm diam. distally with staminate flowers only savaiiense occurs on the north, east, and south and there very slender, filiform, attenuate and slopes of Savai’i in the districts of flexuous, 1.5–2.0 mm diam. proximally with Gagaifomauga III and I, Gaga’emauga III, and triads and later fruits; bracts subtending West Palauli in wet forest and cloud forest from branches not seen, typically caducous. Flowers 900 to 1500 m elevation. I found it in October borne in triads in shallow, cleft-like depressions 2005 at or near the type locality on the steep in proximal one-third to one-half of rachillae, sides of volcanic craters where it occurred as solitary or paired staminate flowers only in scattered individuals emerging from the forest distal one-half to two-thirds of rachillae, each canopy. It is probably scattered in a more or triad subtended proximally by a lip-like less continuous band on the north, east, and bracteole, triads 3–5 mm distant in 2 spiraling south side of the island at the appropriate rows; floral bracteoles 5–6, unequal in size, 3 elevations. Historically, it may have been inner ones broadly triangular to crescent- distributed at lower elevations, perhaps as low shaped, 0.3 × 1.5 mm, imbricate, 3 outer ones as 700 m. The Samoan chief Itutu Avealolo of often more prominent, the 2 outer lateral ones Fogasavi’i, who guided me into the forest to 0.5 × 0.5 mm, tooth-like, outer middle 700 m elevation above Sala’ilua, said it once (proximal) one especially conspicuous and occurred at this location, but tropical cyclones exceeding the triad bracteole, triangular, 1 × and human activity, primarily land clearing 1.5 mm with prominent acute tip, white. and cutting of the trunk for wood, had Staminate flowers 4 × 5 mm, white; sepals 3, destroyed all the specimens in the area. distinct, 2.5–2.8 × 0.4–0.5 mm, narrowly triangular, long-acuminate, margins 3. Clinostigma warburgii Becc., Atti Soc. Tosc. transparent and membranous, briefly connate Sci. Nat. Pisa Mem. 44:155. 1934. Type: or imbricate in basal 0.5 mm; petals 3, distinct, Samoa, ‘Upolu, Lake Lanoto’o, Reinecke 322 5 × 1.5 mm, valvate, widely spreading, free (holotype WRSL?; isotypes BISH, FI). nearly to base, long-acuminate, lanceolate, Clinostigma sp. “Eastern ‘Upolu” Hodel, Palm strongly ribbed when dry; stamens 6, 2.5–3.0 J. 183: 12. mm high, filaments distinct, 1.5–2.0 mm long, very slender, anthers 1 mm long, attached in Solitary, gregarious, unarmed, pleonanthic, middle; pistillode columnar, 1.0–1.1 mm long. monoecious, forest palm to 20 m tall (Front Pistillate flowers 4 × 3.5 mm, ovoid; calyx 3- Cover). Trunk erect, green with white waxy lobed, sepals 3 × 3 mm, faintly ribbed, glaucous indument, aging to grayish white or imbricate nearly to broadly rounded apex; brownish, ringed, to 25 cm diam., internodes petals 3 × 3.5 mm, imbricate nearly to to 10 cm. Leaves 15–20, ascending to

25 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007

10. Clinostigma savaiiense, infructescence, Matavanu, Savai’i, Samoa spreading; sheath 2 m long, lime green with with unbranched portion 10 cm long, rachis slight glaucous bloom, forming a conspicuous of second order branches 12 cm long; rachillae and prominent crownshaft; petiole 50 cm to 30 cm long, white in flower, light green in long, concave adaxially, rounded abaxially; fruit, diffuse, spreading from rachises at angles rachis to 3 m long, flat adaxially, rounded of 45–90 degrees, 0.8-1 mm diam. distally with abaxially; pinnae up to 75 per side, elegantly staminate flowers only and there very slender, pendulous, regularly arranged, lower middle filiform, attenuate and flexuous, 1.5–2.0 mm the largest, to 75 × 4.0–4.5 cm, proximal to 45 diam. proximally with triads and later fruits; × 1.5 cm, distal to 25 × 1 cm, prominently 3- proximal branches subtended by brown, long- nerved adaxially, tan medifixed ramentae to 1 lanceolate bracts to 50 cm long, typically cm long on all 3 nerves abaxially near rachis, caducous, more distal branches and rachillae otherwise glabrous. Inflorescences 9, subtended by rudimentary bracts 1–3 cm long. infrafoliar, to 1.7 × 1.1 m, branched to 3 orders, Flowers borne in triads in shallow, cleft-like most proximal branches the largest and most depressions in proximal one-third to one-half complex, branches becoming progressively of rachillae, solitary or paired staminate flowers smaller and less complex distally until most only in distal one-half to two-thirds of distal represented by simple rachillae; peduncle rachillae, each triad subtended proximally by 20 cm long, 22 cm wide at base and ± bulbous a lip-like bracteole, triads 3-4 mm distant in 2 and swollen, 4 cm wide at first branch, light spiraling rows; floral bracteoles 5-6, unequal in green with slight glaucous bloom and/or size, 3 inner ones broadly triangular to grayish brown scales; prophyll 2 m long, crescent-shaped, 0.75 × 1.75 mm, imbricate, 3 equaling crownshaft, attached 10–12 cm above outer ones often more prominent, the 2 outer base; peduncular bracts 2, attached 2–3 and 6 lateral ones 1 × 0.6 mm, tooth-like, outer cm respectively distally of prophyll middle (proximal) one especially conspicuous attachment, not seen, typically caducous; and exceeding the triad bracteole, broadly rachis 1.2 m long with simple rachillae distally triangular, 1.0–1.3 × 2.0–2.5 mm with and branches proximally, most proximal first prominent acute apex, white. Staminate order branches with unbranched portion 13 flowers 4 × 5 mm, white; sepals 3, distinct, cm long, rachis of first order branches 64 cm 2.0–2.5 × 0.3 mm, narrowly triangular, long- long; most proximal second order branches acuminate, membranous toward apex, margins

26 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007 transparent, briefly connate basally; petals 3, Whistler 3870 (BISH, HAW); NE of Sa’agafou, distinct, 3.5–4.0 × 1–2 mm, valvate, widely 100 m elev., 23 Aug. 1978, Whistler 3923 spreading, free nearly to base, strongly ribbed (HAW); 0.6 miles E. of road into ‘O Le Pupu when dry; stamens 6, 2.0–2.5 mm high, Pu’e National Park, 20 m elev., 25 July 1979, filaments distinct, 1.7–2.5 mm long, very Teraoka & Kennedy 88 (BISH); ‘O Le Pupu Pu’e slender, anthers 0.7–1.3 mm long, attached in National Park, 50 m elev., 23 May 1979, Cox middle; pistillode conical to columnar, 0.7-2 162 (BISH); 40 m elev., 15 March 1980, Moore mm high, to 0.75 mm wide at base, 0.5 mm 10540 (BH); upper entrance to cave at wide at briefly 3-parted apex. Fresh pistillate Togitogiga, 175 m elev., 13 Sept. 1978, Whistler flowers 3–4 × 3–4 mm, ovoid, greenish white; 4004 (BISH, HAW); E of Ti’avea, 220 m elev., calyx 1.5–2.0 × 3 mm, 3-lobed, sepals faintly 24 April 1979, Whistler 4196 (BISH, HAW); ribbed, imbricate in proximal half to two- mangrove swamp at Mulivai, 28 Aug. 1991, thirds, broadly rounded distally with Whistler 8134 (HAW); Puntaemo’o swamp, 31 membranous nearly transparent margins and acute apex; corolla 2.8–3.5 × 3 mm, petals Aug. 1991, Whistler 8159 (HAW); Punataemo’o, imbricate nearly to mucronate apex, whitish, 300 m elev., 14 Nov. 2001, Whistler 11569 thin-fleshy nearly transparent, ribbed; pistil 3 (HAW); Lemafa, 360 m elev., 18 March 1980, × 2.8 mm, ovoid, greenish, 3-parted. Fruits 10 Moore 10541, 10542 (BH); above Sauniatu × 10 mm, rounded, black, stigmatic remains inland from Salafuata, 200–500 m elev. 27 July slightly distal of middle; perianth 3 × 5–6 mm, 1971, Moore 9983 (BH); Apia-Si’umu Road, 450 cupular; sepals 3 × 2 mm, imbricate nearly to m elev. 7 July 1968, Bristol 2179 (BISH); acute apex, broadly rounded to triangular with Afiamalu, 510 m elev., 24 Sept. 1991, Whistler acute apex, petals 4 × 3 mm, imbricate nearly 8387 (HAW); Lake Lanoto’o, Reinecke 322 to briefly acute apex, broadly rounded; (isotype BISH, FI). CULTIVATED: American staminodes 6, 1 × 0.2 mm, tooth-like. Samoa, Tutuila, W. of Aloau (A’oloaufou?), 12 March 1980, Moore 10539 (BH); Ili’ili Village, Specimens Examined. SAMOA. ‘Upolu: 20 m elev., 10 March 1980, Moore 10538 (BH). between Poutasi and Si’umu, 30 m elev., 27 U.S.A., Hawaii, Oahu, Ho’omaluhia Botanic July 1977, Whistler 3858 (BISH, HAW); between Garden (Honolulu Botanic Gardens 78.0889, Falelatai and Lefaga, 150 m elev., 28 July 1977, originally collected by D.R. Hodel, 9 Dec. 1978,

11. Clinostigma warburgii, gregarious population, east of Lemafa Pass, ‘Upolu, Samoa.

27 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007 near Salelesi, ‘Upolu, Samoa, 10 m elev., Hodel Universitatis Florentinae (FI), Museo Botanico, 462), 22 March 2006, Hodel 2008 (BISH). Universita Degli Studi di Firenze, Florence, Distribution and Ecology. Clinostigma warburgii Italy. Robert Dirig, L.H. Bailey Hortorium (BH), primarily occurs on the eastern part of ‘Upolu Cornell University, Ithaca, NY sent loans for in the districts of East Anoama’a, Va’aofonoti, study. Clyde Imada, Bishop Museum; Rusty Aleipata, Lepa, Lotofaga, Falealili, and Si’umu, Russell, Smithsonian Institution (US), from sea level to about 600 m elevation in wet, Washington, D. C.; and Art Whistler and Mike usually disturbed forest (Front Cover). Indeed, Thomas, University of Hawaii herbarium I referred to it in an earlier paper as Clinostigm (HAW), answered questions about the sp. “Eastern ‘Upolu” (Hodel 2006). It is existence of specimens at their institutions. especially abundant in the Lemafa Pass region Jonel Smith, Keith Nobriga, and Joshlyn Sand, and to the east towards Ti’avea, where it forms Ho’omaluhia Botanical Garden, Honolulu, HI, vast, gregarious stands on steep, well drained provided assistance in identifying and slopes and in low, wet or swampy, poorly collecting material of cultivated Clinostigma at drained areas (Fig. 11). It has also been their garden. Audrey and Philip Keeler and the collected twice in the central highlands of Carl and Roberta Deutsch Foundation partially ‘Upolu, once at Afiamalu, where it was supported my field work in the South Pacific. probably cultivated, and at Lake Lanoto’o, the LITERATURE CITED latter represented by Reinecke’s type specimen. BECCARI, O. 1910. Palme Australische nuove o Clinostigma warburgii occurs sparingly farther poco note. Webbia 3: 131–165. west along the north coast in the districts of West Anoama’a and along the south coast in BECCARI, O. 1913 (1914). Manipolo de palme the districts of Safata, Gaaga’emauga, Lefaga, nuove Polinesiane. Webbia 4: 253–291. and Samatau and Falelatai. In these areas, BURRET, M. 1928. Beiträge zur Kenntnis der though, it is typically found as a few isolated Palmen von Malesia, Papua und der Südsee. individuals around human habitation, Repert. Spec. Nov. Regni Veg. 24: 252–296. indicating it is probably cultivated. BURRET, M. 1935a. New palms from Fiji. Occas. When Cox and Moore (1986) concluded that Pap. Bernice P. Bishop Mus. 11(4): 3–14. fruit shape was variable and could not be used to distinguish Clinostigma samoense and C. BURRET, M. 1935b. Palmae gerontogeae IV. onchorhynchum, the specimens from Lemafa Notizbl. Bot. Gart. Berlin-Dahlem 12: Pass that formed the basis for their 590–602. conclusions, Moore 10541 and 10542, actually CHRISTOPHERSEN, E. 1935. Flowering Plants of were C. warburgii. Furthermore, fruits of both collections were immature and at different Samoa. Bernice P. Bishop Mus. Bull. 128: stages of development. When fully mature 1–221. they would likely be the same shape and size. COX, P.A. AND H.E. MOORE, Jr. 1986. The identity The palm illustrated in Langlois (1976, fig. 50, of Clinostigma onchorhynchum. Principes 30: p. 48), captioned as Clinostigma samoense, is 77–81. actually C. warburgii. HODEL, D.R. 1999. Palms for southern Acknowledgments California, Part 36: Clinostigma. Palm J. 144: 11–12. I gratefully acknowledge the assistance several individuals. Michael Grayum, Missouri HODEL, D.R. 2006. Clinostigma, the quint- Botanical Garden (MO), St. Louis, MO, essential palms from the South and West reviewed the manuscript, and he and John Pacific. Palm J. 183: 8–13. Dransfield, Royal Botanic Gardens, Kew (K), LANGLOIS, A.C. 1976. Supplement to Palms of London, England, provided guidance about the World. The University Presses of Florida, nomenclatural matters, especially as they pertained to Clinostigma warburgii. Images of Gainesville, FL. type specimens were provided by Napua MARTELLI, U. 1934. Generi, species e varieta Harbottle, Bishop Museum (BISH), Honolulu, nuove de palme gerontogee della tribu HI; William Baker, Kew; Walter Kittredge, Gray lasciate inedite dal Dr. Od. Beccari Herbarium (GH), Harvard University, MA; and ed ordinate a cura di U. Martelli. Atti Soc. Piero Cuccuini and Egildo Luccioli, Herbarium Tosc. Sci. Nat. Pisa Mem. 44: 114–176.

28 PALMS Hodel: Samoan Clinostigma Vol. 51(1) 2007

MARTELLI, U. 1935. La sinonimia delle palme schaftlichen Forschungsreise nach den gerontogee della tribu delle . Nuovo Samoa-Inseln, dem Neuguinea-Archipel und Giorn. Bot. Ital. 42: 17–87. den Salomons-Inseln, Teil III. Denkschr. Kaiserl. Akad. Wiss., Wien. Math-Naturwiss. MOORE, H.E., Jr. 1969. New palms from the Kl. 85: 1–388. Pacific II. Principes 13: 67–76. WARBURG, O. 1898. Palmae, pp. 588–592 in: F. MOORE, H.E. JR. AND F.R. FOSBERG. The palms of REINECKE, Die Flora der Samoa-Inseln, II. Bot. and the . Gentes Jahrb. Syst. 25: 578–708. Herb. 8: 423–478. RECHINGER, K. 1907. Plantae novae pacificae. WENDLAND, H. 1862. Beiträge zur Palmenflora Repert. Spec. Nov. Regni Veg. 4: 229–233. der Südeeinseln. Bonplandia 10: 190–200.

RECHINGER, K. 1910. Botanische und WHISTLER, W.A. 1992. The palms of Samoa. zoologische Ergebnisse einer wissen- Mooreana 2(3): 24–29.

CLASSIFIED

PERMANENT BOTANICAL GARDEN SIGNS FOR PRIVATE OR PUBLIC COLLECTIONS. Call or write for brochure. Phone: (760) 723-1354; Fax: (760) 723-3220; e-mail: . Gary Wood, PLANT SIGNS, 960 El Caminito, Fallbrook, CA 92028. Web Page: http://www.plantsigns.com

PALM / CYCAS SEEDS We sell RARE and UNCOMMON PALM / CYCAS seeds from all over the world. Seeds from Madagascar, New Caledonia, Bolivia, Seychelles, Solomon Islands, Lord Howe Island and most other countries – including seeds of Coco-de-mer, the infamous Double Coconut. We stock and sell over 300 species from over 40 countries. We supply any quantity. No quantity is too small and none too big. Fresh and viable seeds only. We also carry rare Pachypodium and Adansonia seeds from Madagascar. Not to mention Victoria cruziana and its Longwood hybrid. For more details – please visit our website at http://www.ortanique.com or email us at [email protected] or fax us at 510 494 0105 or write to us at Ortanique, 35314, Rutland Court, Newark, CA 94560, USA.

29